Question: Refer to Example, "Dull Rats," which was done by using a completely randomized design.

a. What were the treatments in this experiment?

b. Were the experimental units the 60 individual rats or the 12 individual experimenters? Explain.

c. Explain how the experiment could have been done using a matched-pair design instead.

Example: Dull Rats In a classic experiment that was designed to test whether the experimenter's expectations could really influence the results, Rosenthal and Fode (1963) deliberately conned 12 experimenters. They gave each one five rats that had been taught to run a maze. They told six of the experimenters that the rats had been bred to do well (i.e., they were "maze bright") and told the other six that their rats were "maze dull" and should not be expected to do well.